Combined Optimization of Trajectory and Design for Dual Propulsion Spacecraft AE5810 Thesis Space Domas M
Total Page：16
File Type：pdf, Size：1020Kb
Combined Optimization of Trajectory and Design for Dual Propulsion Spacecraft AE5810 Thesis Space Domas M. Syaifoel Delft University of Technology Combined Optimization of Trajectory and Design for Dual Propulsion Spacecraft AE5810 Thesis Space by Domas M. Syaifoel to obtain the degree of Master of Science at the Delft University of Technology, to be defended publicly on 14 December 2020. Student number: 4436024 Project duration: November 2019 – December 2020 Thesis committee: Dr. A. Cervone, TU Delft, supervisor Dr. J. Guo TU Delft, committee chair ir. R. Noomen TU Delft An electronic version of this thesis is available at http://repository.tudelft.nl/. Abstract Following the trend of miniaturization and standardization of satellite design, as well as recent successes in the use of cubesats in interplanetary missions, a cubesat design capable of reaching another planet fully in- dependently may lead to significant cost reductions for future missions. While efficient low thrust propulsion systems exist, Earth escape using low thrust only leads to significant incident radiation doses when crossing the Van Allen belts. As such, this report aims to present the design of a dual thrust cubesat, i.e. one which employs both high thrust and low thrust propulsion systems, such that the transfer time and the number of van Allen belt crossings is within requirements, while a final orbit around Mars remains attainable. In doing so, this report explores the theory behind low thrust trajectory optimization, and attempts to combine this with a general optimization scheme including both high thrust phases as well as the optimiza- tion of the spacecraft system design. Implementation of such a scheme has not been shown to be useful: issues in finding a good initial guess prevent convergence in many cases, and the proposed plan to alleviate the encountered issues requires an iterative approach between system design and trajectory design. As such, the proposed scheme has not been found to have any benefit over using existing trajectory design tools in an iterative way. A design approach is presented using the pykep trajectory optimizer developed at the European Space Agency (ESA). Iteration between this tool, high thrust calculations, and system design budget calculations allows for optimizing towards a feasible design that meets the requirements, starting from an arbitrary initial guess. When a list of commercial off the shelf (COTS) components is available following the CubeSat standard, it is possible to quickly generate a feasible design without significant prior work. This report presents a full COTS design for a cubesat capable of reaching Mars independently from a common piggyback launch orbit. The design in question is a 12 unit cubesat with a dry mass of 11.4 kg and a wet mass of 14.8 kg. Launched into a geostationary transfer orbit, it can reach escape velocity though 9 Van Allen belt passes, and reach Mars in less than 5 years. Due to the constrained scope of the design, further work is needed to verify that the momentum dumping budget, link budget, and thermal budget are indeed sufficient. It is further recommended to improve on the work in this thesis, by finding an alternative integrated optimization scheme, or by including and automating more features within pykep. Lastly, recommendations are given for integration, testing, and launch of the proposed spacecraft design. iii Contents Abstract iii Nomenclature vii List of Figures xi List of Tables xiii 1 Introduction 1 2 Literature Study 3 2.1 Motivation............................................3 2.2 Systems Design..........................................3 2.3 Trajectory Design.........................................4 2.3.1 Edelbaum’s Approximation................................4 2.3.2 Direct Method.......................................5 2.3.3 Indirect Method......................................5 2.3.4 Evolutionary Neurocontrol.................................6 2.4 Dual Propulsion Systems.....................................6 2.5 Numerical Aspects........................................7 2.6 Verification & Validation.....................................7 2.7 Conclusion............................................8 3 Design Problem 9 3.1 Intended Achievement......................................9 3.2 Design Objectives.........................................9 3.3 Design Requirements....................................... 10 3.4 Derived Requirements...................................... 11 3.5 Scope............................................... 11 3.6 Design Process Overview..................................... 11 3.7 Using a BVP or OCP Solver.................................... 12 3.8 Using a Trajectory Optimizer................................... 12 3.9 Conclusions............................................ 13 4 Theoretical Approach 15 4.1 State............................................... 15 4.2 Accelerations........................................... 15 4.3 State Changes........................................... 16 4.4 Static Control........................................... 17 4.5 Hamiltonian........................................... 17 4.6 Optimal Dynamic Control.................................... 17 4.7 Costates.............................................. 18 4.7.1 Derivation for Earth Orbits................................. 19 4.7.2 Derivation for Interplanetary Trajectories......................... 19 4.8 Other Costates.......................................... 20 4.9 Conclusions............................................ 21 5 Implementation 23 5.1 Implementation using a BVP Solver................................ 23 5.1.1 Automation of Derivations................................. 24 5.1.2 Cost Function....................................... 25 5.1.3 Differential Equation Function............................... 25 5.1.4 Boundary Condition Function............................... 27 v vi Contents 5.1.5 Elimination of Variables.................................. 27 5.1.6 Initial Guess and Coordinate Systems........................... 28 5.1.7 Initial Guess for Costate Values.............................. 31 5.1.8 Further Issues....................................... 32 5.1.9 Conclusion......................................... 32 5.2 Implementation using an OCP Solver............................... 32 5.2.1 System Definition..................................... 33 5.2.2 Issues........................................... 33 5.3 Implementation using a Trajectory Optimizer.......................... 33 5.3.1 Custom Problem Object.................................. 34 5.3.2 Manual Iteration...................................... 35 5.4 Conclusion of Implementation Attempts............................. 35 6 Resultant Design 37 6.1 Overview............................................. 37 6.2 Initial Guess............................................ 37 6.2.1 High Thrust........................................ 38 6.2.2 Low Thrust......................................... 38 6.2.3 Launch Orbit........................................ 38 6.2.4 Momentum Dumping................................... 39 6.2.5 Solar Array Configuration................................. 39 6.2.6 Initial Budget....................................... 39 6.3 Initial Feasible Trajectory..................................... 40 6.4 Initial Feasible Combined Design................................. 41 6.5 Initial Feasible COTS Design................................... 41 6.6 Current Design.......................................... 43 6.6.1 On Board Computer.................................... 43 6.6.2 Electric Power System................................... 44 6.6.3 Communication...................................... 44 6.6.4 Attitude Determination and Control............................ 44 6.6.5 Solar Array......................................... 48 6.6.6 Electric Thruster...................................... 48 6.6.7 Chemical Thruster..................................... 49 6.6.8 Radiation......................................... 49 6.6.9 Interface Layout...................................... 49 6.6.10 Spacecraft Modes..................................... 50 6.6.11 Alternatives........................................ 51 6.7 Design Recommendations.................................... 51 6.8 Verification and Validation.................................... 51 6.8.1 Requirement Validation.................................. 53 6.8.2 System Verification..................................... 53 6.8.3 System Validation..................................... 54 6.8.4 Conclusions........................................ 54 7 Conclusion and Recommendations 55 Bibliography 59 Nomenclature Abbreviations ADC Attitude determination and control subsystem AU Astronomical unit BVP Boundary value problem CAD Computer aided design COM Communication subsystem COTS Commercial off the shelf CPT Chemical thruster propellant subsystem CTH Chemical thruster subsystem CTR Structural subsystem EPR Electric thruster propellant subsystem EPS Electric power subsystem ESA European Space Agency ETH Electric thruster subsystem GTO Geostationary transfer orbit I2C Interintegrated circuit LEO Low Earth orbit MBH Monotonic basin hopping MPPT Maximum power point tracker NASA National Aeronautics and Space Administration OBC On board computer subsystem OCP Optimal control problem ODE Ordinary differential equation PCA Patched conic approximation PMP Pontryagin’s maximum principle RAD Radiation shielding subsystem SOI Sphere of influence SOL Solar array subsystem